Eureka 2 has been discontinued

[Julien Dubois]

Hi everyone,

So the news is here, Eureka 2 has been discontinued: https://github.com/Netflix/eureka/wiki#eureka-20-discontinued

As of today we have typically 3 stacks, and I just wanted to tell you my current analysis on this, and get your feedback:

- A Spring Cloud + Netflix stack -> this is our most widely used stack, which is very stable and which I would recommend. But this is clearly dying: Eureka is dying, and now the Spring Cloud team is pushing Spring Cloud gateway against Zuul 2. Please note that I'm not a big fan of implementations by the Spring team: in the past, most of them have been failures, so I'd rather stick on using Spring as a framework, but nothing more. Anyway, fragmentation here means also smaller market shares, and this is bad as Open Source projects only succeed when they have big market share.

- A Spring Cloud + Consul / Traefik stack -> this is kind of an hybrid stack. This is good as this is "cloud neutral", and a lot of people use Consul.

- An Istio stack -> basically this gets rid of a lot of Spring Cloud features. Clearly this is where the hype is today, and also the Google money. I don't think Consul will disappear easily (and they are providing some Istio-like features now), but if I needed to bet on something, it would be on this stack. The main downside is that this takes many configurations we were doing on the developer side to put them on the production side - no wonder that Google pushes this, as this means more people will host their architecture on GCP.

Whatever happens, we support the 3 stacks, so that's not a big issue for us. But we need to put our effort where it's most useful. What do you think of this?

Cheers,

Julien

[Deepu K Sasidharan]

Agree with you. My few cents

I think the first stack is the simplest to use and understand and hence easier for beginners and intermediate developers. What options do we have to support this since Eureka is dying? is there something else that we can migrate JHipster registry to?

For option 2 and 3 we need them but they are harder to understand and work with (at least at the moment)

Also from our stats which is the most popular usage?

Thanks & Regards,

Deepu

--
You received this message because you are subscribed to the Google Groups "JHipster dev team" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jhipster-dev...@googlegroups.com.
To post to this group, send email to jhipst...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jhipster-dev/CADNXADGmcjTcv-EmSnfSy%3D4EtfED9R2T-NS1U5-jxC0kjp37yw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[Sendil Kumar N]

I too agree with you. It is sad to see Eureka dying. Zuul2 promises a lot but it may also go away.

Spring cloud gateway is in its early stage but it is promising. At one point or the other, people will definitely start asking about it. 

I think in the future, it is better to move smaller market share items into blueprint and make them separate projects. This will make our application light weight.

I personally like Istio, because it takes away most of the things. I do see a lot of people love Consul / Traefik. It gives the flexibility and the variety for people to choose.

Cheers

Sendil Kumar

To view this discussion on the web visit https://groups.google.com/d/msgid/jhipster-dev/CAFXry57LDOHgtU0ny339ov3kLrEv6DWiNjLewgbTq23KafhQaw%40mail.gmail.com.

[Gaël Marziou]

I really like Consul from a technical standpoint but licence fees for support is soooooo expensive that I stayed with Eureka.

So I'd bet on Istio too or on next big thing :)

Gael